Exact Bayesian structure learning from uncertain interventions

By Daniel Eaton, Kevin P. Murphy. 2007

233 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods, Causal Inference & Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

In Causal Inference & Discovery

Citations in the corpus

Disentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICA (2021, background): “…has been explored previously in the context of causal discovery where the intervention occurs on observed variables instead of latent variables like in our case (Eaton and Murphy, 2007; Mooij et al., 2020; Squires et al., 2020; Jaber et al., 2020; Brouillard et al., 2020; Ke et al., 2019).”
Differentiable Causal Discovery from Interventional Data (2020, background): “The general type of interventions described in (2) are called imperfect (or soft, parametric) [36, 7, 8].”
Toward Causal Representation Learning (2021, background): “Several types of interventions may be possible [63], which can be categorized as follows.”
Towards Causal Representation Learning (2021, background): “Several types of interventions may be possible [62] which can be categorized as: No intervention: only observational”
Learning Neural Causal Models from Unknown Interventions (2019, methodology): “Eaton & Murphy (2007b) use dynamic programming techniques to accelerate Markov Chain Monte Carlo (MCMC) sampling in a Bayesian approach to structure learning for discrete variable DAGs.”
Invariant Causal Prediction for Sequential Data (2017, methodology): “ral. Having access to interventional data improves idenitiﬁability for inferring causal relations and several methods have been proposed that exploit both observational and interventional data (e.g., Eaton and Murphy, 2007; Hauser and Bühlmann, 1 2012; Peters et al., 2016). Causal discovery, however”
Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks (2014, background): “Different types of “imperfect” interventions can be considered as well, see e.g., Eberhardt and Scheines (2007); Eaton and Murphy (2007); Mooij and Heskes (2013).”
Characterization and greedy learning of interventional Markov equivalence classes of directed acyclic graphs (2011, methodology): “The Bayesian procedures of Cooper and Yoo (1999) or Eaton and Murphy (2007) address the problem of calculating a posterior (and also a likelihood) of an ensemble of observational and interventional data but do not address questions of identiﬁability or Markov equivalence: allowing diﬀerent…”
Causal Structure Learning (2017, methodology): “Other excluded methods that make use of interventional data include Cooper and Yoo (1999); Tian and Pearl (2001) and Eaton and Murphy (2007), where the latter does not require knowledge of the precise location of interventions in a similar spirit to Rothenh¨ausler et al. (2015).”
Network Inference and Biological Dynamics (2011, background): “Existing literature in causal inference [Pearl (2009)] and related work in graphical models [Eaton and Murphy (2007)] are relevant, but in biological applications it may also be important to consider the mechanism of action of speciﬁc interventions.”
Modeling Discrete Interventional Data using Directed Cyclic Graphical Models (2009, methodology): “While we have assumed that the interventions are ‘perfect’, in many cases it might be more appropriate to use DCG models with ‘imperfect’, ‘soft’, or ‘uncertain’ interventions (see Eaton and Murphy (2007)).”
Permutation-Based Causal Structure Learning with Unknown Intervention Targets (2019, methodology): “We showed that the I -Markov equivalence class is identiﬁable even without prior knowledge of the intervention targets, a theoertical result of independent interest (Eaton and Murphy, 2007; Mein-shausen et al., 2016; Rothenh¨ausler et al., 2015; Ghas-sami et al., 2017).”
Learning Bayesian networks: approaches and issues (2011, methodology): “More recently, Eaton and Murphy (2007a) demonstrate a method that utilizes MCMC and dynamic programming (as in Section 4.11) in model averaging.”
Foundations of structural causal models with cycles and latent variables (2016, background): “Other types of interventions, like mechanism changes [72], fat-hand interventions [12], activity interventions [42] and stochastic versions of all these are at least as relevant, but we do not consider them here.”